Textbook Outcomes Among Medicare Patients Undergoing Hepatopancreatic Surgery

Abstract

Objective: To define and test “Textbook Outcome” (TO)—a composite measure for healthcare quality—among Medicare patients undergoing hepatopancreatic resections. Hospital variation in TO and Medicare payments were analyzed. Background: Composite measures of quality may be superior to individual measures for the analysis of hospital performance. Methods: The Medicare Provider Analysis and Review (MEDPAR) Inpatient Files were reviewed to identify Medicare patients who underwent pancreatic and liver procedures between 2013 and 2015. TO was defined as: no postoperative surgical complications, no prolonged length of hospital stay, no readmission ≤ 90 days after discharge, and no postoperative mortality ≤ 90 days after surgery. Medicare payments were compared among patients who achieved TO versus patients who did not. Multivariable logistic regression was used to investigate patient factors associated with TO. A nomogram to predict probability of TO was developed and validated. Results: TO was achieved in 44% (n = 5919) of 13,467 patients undergoing hepatopancreatic surgery. Adjusted TO rates at the hospital level varied from 11.1% to 69.6% for pancreatic procedures and from 16.6% to 78.7% for liver procedures. Prolonged length of hospital stay represented the major obstacle to achieve TO. Average Medicare payments were substantially higher among patients who did not have a TO. Factors associated with TO on multivariable analysis were age, sex, Charlson comorbidity score, previous hospital admissions, procedure type, and surgical approach (all P > 0.05). Conclusions: Less than one-half of Medicare patients achieved a TO following hepatopancreatic procedures with a wide variation in the rates of TO among hospitals. There was a discrepancy in Medicare payments for patients who achieved a TO versus patients who did not. TO could be useful for the public reporting of patient level hospital performance and hospital variation.

Incidence and impact of Textbook Outcome among patients undergoing resection of pancreatic neuroendocrine tumors: Results of the US Neuroendocrine Tumor Study Group

“…Mortality was defined as death within 90 days of the index operation. Readmission was defined as admission to index or nonindex hospital within 90 days after discharge 13 . An R0 resection was classified as a negative margin width > 1 mm, whereas a margin width ≤ 1 mm was categorized as R1.…”

“…In turn, composite outcomes, which reflect an “all‐or‐none” metric, may be more consistent with patient expectations around an optimal hospital course following surgery. One such composite measure is “Textbook Outcome” (TO), which has been used to evaluate outcomes among patients undergoing surgery for colorectal, vascular, esophageal, and more recently, hepato‐pancreato‐biliary disease 10‐15 . In fact, in esophageal cancer, TO has been related to better overall and conditional overall survival (c‐OS) 14 …”

Comparing textbook outcomes among patients undergoing surgery for cancer at U. S. News & World Report ranked hospitals

“…Van Roessel et al 38 noted that TO can better reflect the quality and may be a powerful parameter to assess quality between different hospitals during regular audits for pancreatic surgery. Similarly, several other investigators reported TO as a useful tool to assess interhospital variations among patients undergoing complex surgery 10,11,37 . As the TO composite measure combines important quality parameters into a single metric, TO may increase the reliability to evaluate overall hospital performance.…”

“…TO was defined as no postoperative surgical complications, no prolonged length of hospital stay (LOS), no readmission 90 days after discharge, and no postoperative mortality 90 days after surgery. TO definition was based on a selection of relevant outcome parameters representing an optimal patient outcome after surgery 10 . ICD‐9‐CM diagnosis and procedure codes were utilized to identify patients having postoperative surgical complications 13–15 .…”

“…In addition, parameters with a low event rate, such as mortality, maybe due to chance and focus on such metrics may not provide reproducible and accurate information into actual variation around quality measures among different hospitals. In turn, several investigators have proposed composite outcomes measures as being more reliable to assess healthcare quality rather than focusing on individual outcome parameters 10–12 . Textbook outcome (TO) is one such composite measure that has gained increased adoption as a means to provide more comprehensive information about hospital performance.…”